On October 25, 2017, the FDA published two new guidance documents on drug-drug interactions (DDI). These guidance documents replace the February 2012 guidance, “Drug Interaction Studies—Study Design, Data Analysis, Implications for Dosing, and Labeling Recommendations.” According to the FDA, these new guidances reflect the agency’s current thinking and greater learnings on DDI and provides a more systematic and risk-based approach to this critical topic. Additionally, it creates further alignment with other global regulatory agencies, specifically the EMA and Japan’s PMDA.
- The first guidance, “In Vitro Metabolism- and Transporter-mediated Drug-drug Interaction Studies” addresses how to extrapolate in vitro data to determine if clinical DDI trials are required, and if so, how those data can inform the trials. This decision-making process generally (but not always) occurs early in the drug development process with the potential DDI liability impacting a sponsor’s decision to move forward with an investigational drug. This guidance includes considerations when choosing in vitro experimental systems, key issues regarding in vitro experimental conditions, and more detailed explanations regarding model-based DDI prediction strategies.
- If an in vitro assessment as determined from the above guidance suggests that the sponsor should conduct a clinical DDI study, that sponsor should refer to the second new related guidance, “Clinical Drug Interaction Studies—Study Design, Data Analysis, and Clinical Implications,” which addresses the conduct and interpretation of clinical DDI studies.
Growing Importance of Modeling and Simulation (M&S)
These new guidances demonstrate the FDA’s increased confidence in M&S for drug development and review. In a July 7 announcement regarding the steps the agency is taking to implement the 21st Century Cures Act, Commissioner Scott Gottlieb wrote:
Modeling and simulation play a critical role in organizing diverse data sets and exploring alternate study designs. This enables safe and effective new therapeutics to advance more efficiently through the different stages of clinical trials. FDA’s efforts in modeling and simulation are enabled through multiple collaborations with external parties that provide additional expertise and infrastructure to advance the development of these state-of-the-art technologies. FDA’s Center for Drug Evaluation and Research (CDER) is currently using modeling and simulation to predict clinical outcomes, inform clinical trial designs, support evidence of effectiveness, optimize dosing, predict product safety, and evaluate potential adverse event mechanisms.
The endorsement of M&S for informing DDI risk assessment is also seen in these new guidances. The In Vitro DDI guidance includes a chapter called “Using Model-based Predictions to Determine a Drug’s Potential to Cause DDIs,” which outlines a range of M&S approaches to translate in vitro observations into in vivo predictions of potential clinical DDIs—from basic kinetic models to both static and dynamic mechanistic models that include physiologically-based pharmacokinetic (PBPK) models. In many cases, negative findings from early in vitro and clinical studies, in conjunction with model-based predictions, can eliminate the need for additional clinical investigations of a drug’s DDI potential. PBPK models can predict the DDI potential of an investigational drug as an enzyme substrate or an enzyme perpetrator. Alternatively, the sponsor can use a PBPK model to inform the need for conducting additional studies.
The new clinical DDI guidance speaks to M&S for both informing DDI clinical trials and replacing the need for trials. The sponsor can simulate various DDI scenarios using available pharmacokinetic models (either mechanistic PBPK models or empirical population pharmacokinetic models) to optimize study sampling (eg, sampling times, number of subjects) and data collection. Population pharmacokinetic analyses of data obtained from large-scale clinical studies can help characterize the clinical impact of known or newly identified interactions and determine recommendations for treatment modifications when the investigational drug is a substrate.
The clinical DDI guidance also outlines how PBPK models can be used in lieu of some prospective DDI studies. For example, PBPK models have predicted the impact of weak and moderate index inhibitors on some CYP2D6 and CYP3A substrates as well as the impact of weak and moderate index inducers on CYP3A substrates. The chart below shows the increasing acceptance of this approach in the FDA’s acceptance of PBPK in lieu of clinical trials.
Taken together, these guidances demonstrate the great progress that M&S has made in drug development and regulatory review. Specifically, they clarify the FDA’s growing comfort and reliance on these methods. Further, they speak to the future of M&S within regulatory science and ongoing work to advance its use: “PBPK models can include ADME processes mediated by transporters as well as passive diffusion and metabolism. However, compared to CYP enzymes, the predictive performance of PBPK modeling for transporter-based DDIs has not been established.” We would say, has not yet been established!
To learn how modeling and simulation can help you understand the mechanisms of complex DDIs, please read this case study.